The process of analyzing data from different perspectives and summarizing it into useful information is defined as data mining (knowledge discovery).It allows users to analyze data from many different dimensions or angles, categorize it, and summarize the relationships identified. We focused on bridging the vocabulary gap between health seekers and providers to code the medical records by utilizing local mining and global learning approaches. Generally, the community-based healthcare service including HealthTap2, HaoDF3 and WebMD4 are disseminating personalized health knowledge and connecting patients with doctors worldwide via question answering. It is difficult to avail online always by health service providers (Doctors). Further, most of the previous work simply utilizes the external medical dictionary to code the medical records rather than considering the corpus-aware terminologies which is focused on hospital generated health data or health provider released sources by utilizing either isolated or loosely coupled rule-based and machine learning approaches. We propose a novel scheme that is able to code the medical records with corpus-aware terminologies .Additional to these, we incorporated Separate Server Implementations for Local Mining and Global Learning, Medi Net Library, NLP Techniques, PDF Searching using Indexing and android applications for users.